How the many different cell types of the nervous system arise from a few indifferent precursors, which themselves originally arose from a single pluripotent cell is a major question in developmental biology. In this proposal we address this question through cell lineage studies of the frog embryo (Xenopus laevis): a single cell is identified early in development and its progeny followed as development proceeds. We propose to use magnetic resonance imaging (MRI) microscopy to follow single cells and/or groups of cells in the developing animal. MRI is a non-invasive technique capable of obtaining 3-D images of opaque samples on a timescale of minutes, i.e. the development of a single embryo can be followed in vivo. Membrane impermeable MRI contrast agents (e.g. Gd-DTPA covalently linked to dextran or superparamagnetic iron micro-particles) make ideal lineage tracers because they drastically alter the MRI signal intensity in their local environment. The lineage tracer will be injected into a proliferating cell. Being impermeable to the cell membrane, the tracer will be inherited by all the descendants of the injected cell and allow them to be identified in the magnetic resonance image. MRI has the theoretical resolution and contrast necessary to detect individual daughter cells. Preliminary experiments in our laboratories indicate the practical feasibility of this approach: the contrast agents are effective at low concentrations and non-toxic, and we have obtained magnetic resonance images with 4mum resolution. MRI will be used to obtain 3-dimensional representations of the animal as a function of time after injection. We will be able to locate the position of each labeled cell in three dimensions at a series of time points in the developing embryo with a temporal resolution of less than 1/2 hour. The three dimensional images can be analyzed either retrospectively or prospectively to gain information about issues such as secondary & tertiary lineage branching, cell motion, and cell-cell interactions. A 'movie' consisting of one 3-D image every half-hour over the course of several days to a week will provide significant insight into the exact lineages/motions of cells in the developing animal. Presently, this is not possible with any other technique.